325 results about "Neural network architecture" patented technology

A color management method / apparatus generates image color matching and International Color Consortium (ICC) color printer profiles using a reduced number of color patch measurements. Color printer characterization, and the generation of ICC profiles usually require a large number of measured data points or color patches and complex interpolation techniques. This invention provides an optimization method / apparatus for performing LAB to CMYK color space conversion, gamut mapping, and gray component replacement. A gamut trained network architecture performs LAB to CMYK color space conversion to generate a color profile lookup table for a color printer, or alternatively, to directly control the color printer in accordance with the a plurality of color patches that accurately. represent the gamut of the color printer. More specifically, a feed forward neural network is trained using an ANSI / IT-8 basic data set consisting of 182 data points or color patches, or using a lesser number of data points such as 150 or 101 data points when redundant data points within linear regions of the 182 data point set are removed. A 5-to-7 neuron neural network architecture is preferred to perform the LAB to CMYK color space conversion as the profile lookup table is built, or as the printer is directly controlled. For each CMYK signal, an ink optimization criteria is applied, to thereby control ink parameters such as the total quantity of ink in each CMYK ink printed pixel, and / or to control the total quantity of black ink in each CMYK ink printed pixel.

Systems and methods are disclosed for routing callers to agents in a contact center, along with an intelligent routing system. An exemplary method includes combining multiple output variables of a pattern matching algorithm (for matching callers and agents) into a single metric for use in the routing system. The pattern matching algorithm may include a neural network architecture, where the exemplary method combines output variables from multiple neural networks. The method may include determining a Z-score of the variable outputs and determining a linear combination of the determined Z-scores for a desired output. Callers may be routed to agents via the pattern matching algorithm to maximize the output value or score of the linear combination. The output variables may include revenue generation, cost, customer satisfaction performance, first call resolution, cancellation, or other variable outputs from the pattern matching algorithm of the system.

Systems and methods are disclosed for routing callers to agents in a contact center, along with an intelligent routing system. An exemplary method includes combining multiple output variables of a pattern matching algorithm (for matching callers and agents) into a single metric for use in the routing system. The pattern matching algorithm may include a neural network architecture, where the exemplary method combines output variables from multiple neural networks. The method may include determining a Z-score of the variable outputs and determining a linear combination of the determined Z-scores for a desired output. Callers may be routed to agents via the pattern matching algorithm to maximize the output value or score of the linear combination. The output variables may include revenue generation, cost, customer satisfaction performance, first call resolution, cancellation, or other variable outputs from the pattern matching algorithm of the system.

Provided herein are methods and systems for implementing three-dimensional perception in an autonomous robotic system comprising an end-to-end neural network architecture that directly consumes large-scale raw sparse point cloud data and performs such tasks as object localization, boundary estimation, object classification, and segmentation of individual shapes or fused complete point cloud shapes.

The invention discloses a convolution network arithmetic unit, a reconfigurable convolution neural network processor and an image de-noising method of the reconfigurable convolution neural network processor. The reconfigurable convolution neural network processor comprises a bus interface, a preprocessing unit, a reconfigurable hardware controller, an SRAM, an SRAM control module, an input caching module, an output caching module, a memory, a data memory controller and the convolution network arithmetic unit. The processor is featured by few resources and rapid speed and can be applicable to common convolution neural network architecture. According to the unit, the processor and the method provided by the invention, convolution neural networks can be realized; the processing speed is rapid; transplanting is liable to be carried out; the resource consumption is little; an image or a video polluted by raindrops and dusts can be recovered; and raindrop and dust removing operations can be taken as preprocessing operations for providing help in follow-up image identification or classification.

Fractal computers are neural network architectures that exploit the characteristics of fractal attractors to perform general computation. This disclosure explains neural network implementations for each of the critical components of computation: composition, minimalization, and recursion. It then describes the creation of fractal attractors within these implementations by means of selective amplification or inhibition of input signals, and it describes how to estimate critical parameters for each implementation by using results from studies of fractal percolation. These implementation provide standardizable implicit alternatives to traditional neural network designs. Consequently, fractal computers permit the exploitation of alternative technologies for computation based on dynamic systems with underlying fractal attractors.

A system and method for semantic extraction using a neural network architecture includes indexing each word in an input sentence into a dictionary and using these indices to map each word to a d-dimensional vector (the features of which are learned). Together with this, position information for a word of interest (the word to labeled) and a verb of interest (the verb that the semantic role is being predicted for) with respect to a given word are also used. These positions are integrated by employing a linear layer that is adapted to the input sentence. Several linear transformations and squashing functions are then applied to output class probabilities for semantic role labels. All the weights for the whole architecture are trained by backpropagation.

The invention discloses a sparse neural network architecture and a realization method thereof. The sparse neural network architecture comprises an external memory controller, a weight cache, an inputcache, and output cache, an input cache controller and a computing array, wherein the computing array comprises multiple computing units, each row of reconfigurable computing units in the computing array share partial input in the input cache, and a partial weight, shared by each column of reconfigurable computing units, in the weight cache is computed; the input cache controller performs sparse operation on input of the input cache, and a zero value in the input is removed; and the external memory controller stores data of the computing array before and after processing. Through the sparse neural network architecture and the realization method thereof, invalid computing performed when the input is zero can be reduced and even eliminated, computed quantities among all the computing units are balanced, the hardware resource utilization rate is increased, and meanwhile shortest computing delay is guaranteed.

The invention discloses a multi-target tracking method based on a recurrent neural network, comprising the following steps: building a monitoring video data set marked with each frame of pedestrian location; manually expanding the monitoring video data set marked with each frame of pedestrian location to get training set samples; grouping the training set samples to get multiple training groups; building a multi-target tracking network; inputting the training groups in sequences to the multi-target tracking network for training; and inputting video data to be tested to the trained multi-target tracking network and performing forward propagation to get the moving trajectories of multiple targets. According to the invention, a proposed network model is trained in an end-to-end manner using raw data and a large amount of data obtained through artificial expansion, complex tasks such as data association and track estimation are completed under a unified neural network architecture, and the moving trajectories of targets can be tracked effectively under different directions, light conditions, deformations and other complex environments.

The present invention discloses a sequence recommendation method based on user behavioral difference modeling. The method comprises: acquiring historical behavior information of a user; calculating acommodity feature vector according to the acquired historical behavior information; by combining a commodity feature vector, using a behavioral difference modeling method for sequence modeling, and obtaining the current demand and historical preferences of the user by using two different neural network architectures; and according to the current purchase demand and the historical preferences of the user, predicting the next commodity of interest to the user through joint learning, performing matching in a commodity vector space, finding a plurality of commodities that are closest to the predicted result in the commodity vector space, and generating a commodity recommendation sequence. According to the method provided by the present invention, through difference modeling on user timing sequence behaviors, the current demand and long-term preferences in the purchase decision of the user can be intelligently understood, and accurate sequence recommendation services can be provided for users.

The invention belongs to the technical field of deep learning, and particularly relates to a neural network architecture search method, a neural network application method, equipment and a storage medium. The method comprises the steps of searching a first neural network model through a reinforcement learning architecture search method, and performing network structure search on the first neural network model based on a preset jump connection number to obtain a second neural network model; and setting the structure of the second neural network model based on parameters of a pre-established evaluation network, and performing training on a test data set to obtain a trained neural network model. The neural network architecture in the method is higher in search speed, and the problem that network architecture search resources are consumed too much is effectively solved. The target image is processed through the neural network model obtained through training, various image processing taskscan be executed, and the recognition precision is higher during image processing.

The invention relates to a deep-learning-based classification and grading method for diabetes retinopathy. For each kind of diabetes retinopathy, lots of ophthalmoscope photographs are prepared; a deep convolution neural network including multi-level neural network architecture is established; the deep convolution neural network is trained based on the lots of ophthalmoscope photographs, so that the final output value of the deep convolution neural network meets a grading result of the ophthalmoscope photographs; and thus disease grading can be carried out by using the trained deep convolution neural network. According to the method provided by the invention, with application of the lots of ophthalmoscope photographs including diagnosis marks, needed features are learned automatically from a training example base by deep learning and grading diagnosis is carried out; and data features for determination and the deep convolution neural network parameters are corrected continuously during the training process, so that the grading accuracy and reliability in a real application scene can be improved substantially.

The invention discloses a self-adaption control method and a device based on deep learning. According to the invention, a deep neural network is utilized to learn to realize the online control of people on a back route through the obtaining of feedback of the body of an experienced driver in various driving conditions and to generate a self-adaption controller. The method and the device are applied to self-driving and are suitable for various types of vehicles. In the invention, the cognitive behaviors of the experienced driver are materialized; the self-driving control is decoupled; and the circular neural network architecture in deep learning is used to make a self-adaption controller of a self-driving vehicle. In the application, the self-adaption controller receives the command of the cognitive arrows formed by the driving brain decision so as to deal with the road, weather and load uncertainties to control the driving of the vehicle and to ensure the self-driving becomes safe, stable and energy conserving.

The present invention provides an aircraft engine vibration system that provides information about engine health. Embodiments of the present invention monitor for excessive vibration, monitor for bird strike, monitor for ice build up on the fan section, and monitor general engine health. An embodiment of the present invention utilizes neural network architecture for the detection of excessive vibration and ice detection build-up on the fan section of a turbo-fan engine and to monitor engine health through the high-pressure turbine section of the engine.

Provided herein are methods and systems for implementing three-dimensional perception in an autonomous robotic system comprising an end-to-end neural network architecture that directly consumes large-scale raw sparse point cloud data and performs such tasks as object localization, boundary estimation, object classification, and segmentation of individual shapes or fused complete point cloud shapes.

A technique for machine learning, such as supervised artificial neural network learning includes receiving data and checking the dimensionality of the read data and reducing the dimensionality to enhance machine learning performance using Principal Component Analysis methodology. The technique further includes specifying the neural network architecture and initializing weights to establish a connection between read data including the reduced dimensionality and the predicted values. The technique also includes performing supervised machine learning using the specified neural network architecture, initialized weights, and the read data including the reduced dimensionality to predict values. Predicted values are then compared to a normalized system error threshold value and the initialized weights are revised based on the outcome of the comparison to generate a learnt neural network having a reduced error in weight space. The learnt neural network is validated using known values and is then used for predicting values.

A deep approximation neural network architecture which extrapolates data over unseen states for demand response applications in order to control distribution systems like product distribution systems of which energy distribution systems, e.g. heat or electrical power distribution, are one example. The method is a model-free control technique mainly in the form of Reinforcement Learning (RL) where a controller learns from interaction with the system to be controlled to control product distributions of which energy distribution systems, e.g. heat or electrical power distribution, are one example.

The invention relates to a generative adversarial network method fusing a self-attention mechanism, belongs to the field of computer vision, and particularly relates to a generative adversarial network for image generation. The generation of the image is an important challenge in the field of computer vision, and if a large number of high-quality image samples can be generated, the artificial intelligence field can be more rapidly developed in the era depending on the big data background. Therefore, the invention provides the generative adversarial network fusing the self-attention mechanism,the network can generate the high-quality images, and meanwhile, the images have higher diversity. Specifically, the generative adversarial network uses a Waserstein distance to measure an evaluationcriterion of the generator and discriminator distribution, and a loss function is correspondingly improved; meanwhile, the self-attention mechanism is introduced into the neural network architecture corresponding to the generator and the discriminator, so that the relevance between the local pixel regions of the generated image is improved, and the quality of the generated image is improved.

A system for visual odometry is disclosed. The system includes: an internet server, comprising: an I / O port, configured to transmit and receive electrical signals to and from a client device; a memory; one or more processing units; and one or more programs stored in the memory and configured for execution by the one or more processing units, the one or more programs including instructions for: extracting representative features from a pair input images in a first convolution neural network (CNN) in a visual odometry model; merging, in a first merge module, outputs from the first CNN; decreasing feature map size in a second CNN; generating a first flow output for each layer in a first deconvolution neural network (DNN); merging, in a second merge module, outputs from the second CNN and the first DNN; generating a second flow output for each layer in a second DNN; and reducing accumulated errors in a recurrent neural network (RNN).

Systems, methods and computer-readable medium for (i) accelerating the inference speed of a deep neural network (DNN), and (ii) compressing the vector representations produced by the DNN out of a variety of input data, such as image, audio, video and text. A method embodiment takes as inputs a neural network architecture and a task-dependent loss function, measuring how well a neural network performs on a training data set, and outputs a deep neural network with sparse neuron activations. The invented procedure augments an existing training objective function of a DNN with regularization terms that encourage sparse activation of neurons, and compresses the DNN by solving the optimization problem with a variety of algorithms. The present disclosure also shows how to utilize the sparsity of activations during the inference of DNNs so the number of arithmetic operations can be reduced proportionately, and how to use the sparse representations produced by the DNNs to build an efficient search engine.

The present invention relates to a BDPD-based method for improving efficiency of RF power amplifier, comprising: first, choose key neural network architecture and scale and input initial values of modeling data and network parameters necessary for establishing the neural network model for RF power amplifier; second, correct network parameters with back propagation method and output the neural network model for RF power amplifier when the error meets the criterion; next, solve the pre-distortion algorithm of the RF power amplifier with said model and then carry out pre-distortion processing for the input with the pre-distortion algorithm and feed the input to the RF power amplifier. The present invention can be used to establish a neural network model with adequate accuracy and easy to solve corresponding pre-distortion algorithm for RF power amplifier, in order to improve RF power amplifier efficiency, reduce costs, and suppress out-of-band spectrum leakage effectively through base-band digital pre-distortion technology.

A method for stimulation of hydrocarbon production via imbibition by utilization of surfactants. The method includes use of fuzzy logic and neural network architecture constructs to determine surfactant use.

The invention relates to a neural network structure searching method and system, a storage medium and equipment, and the method comprises the steps: S1, obtaining a preset neural network architectureand a sampling structure; wherein the neural network architecture comprises an input layer, an output layer and a plurality of intermediate layers which are arranged in sequence; S2, according to thenetwork structure to be determined in each intermediate layer and the corresponding structure search interval, sampling for multiple times through a sampling structure, and obtaining a plurality of sub-neural-network structures; S3, performing classification training on the plurality of sub-neural-network structures to obtain a plurality of updated sub-neural-network structures and updated structure search intervals of the intermediate layers; s4, obtaining the classification accuracy of the updated plurality of sub-neural network structures, and updating the parameters of the sampling structure according to the accuracy; and S5, determining a required neural network structure. According to the method, the neural network structure with the effect most matched with the classification task can be automatically searched, time is saved, and efficiency is improved.

Fractal computers are neural network architectures that exploit the characteristics of fractal attractors to perform general computation. This disclosure explains neural network implementations for each of the critical components of computation: composition, minimalization, and recursion. It then describes the creation of fractal attractors within these implementations by means of selective amplification or inhibition of input signals, and it describes how to estimate critical parameters for each implementation by using results from studies of fractal percolation. These implementation provide standardizable implicit alternatives to traditional neural network designs. Consequently, fractal computers permit the exploitation of alternative technologies for computation based on dynamic systems with underlying fractal attractors.

A method for stimulation of gas hydrocarbon production via imbibition by utilization of surfactants. The method includes use of fuzzy logic and neural network architecture constructs to determine surfactant use.

The invention relates to a construction method and application of a lightweight gesture detection convolutional neural network model, and the method comprises the steps: constructing a lightweight gesture detection convolutional neural network framework based on a SquezeNet convolutional neural network framework and an SSD multi-target detection convolutional neural network framework; Acquiring agesture picture and a background picture, and performing image data enhancement and picture synthesis processing on the gesture picture based on the background picture to obtain a gesture data set; And based on the public data set and the gesture data set, training a lightweight gesture detection convolutional neural network framework to obtain a lightweight gesture detection convolutional neuralnetwork model. According to the invention, a small amount of gesture data is expanded into the gesture data set containing a large amount of picture data at a high speed; The technical problem that alarge amount of high-quality gesture picture data is difficult to obtain is solved, in addition, by combining the SquezeNet convolutional neural network architecture and the SSD multi-target detectionconvolutional neural network architecture, the constructed lightweight gesture detection convolutional neural network model occupies few computing resources, and can be applied to various detection platforms.

The invention discloses a neural network architecture evaluation method based on attribute graph optimization, and the method comprises the steps: modeling a neural network architecture as an attribute graph, and constructing a Bayesian graph neural network agent model; randomly generating, training, and testing a group of neural network architectures, taking the group of neural network architectures and performance indexes corresponding to testing as an initial training set, wherein the training set is used for training a Bayesian graph neural network agent model; according to the current training set, generating a new neural network candidate set through an evolutionary algorithm and training a Bayesian graph neural network agent model; selecting a potential individual from the neural network candidate set by maximizing a collection function, then training and testing the individual, and adding the individual and a performance index corresponding to the test into the current training set; and under the constraint of fixed cost, repeating the above steps until the best neural network architecture and the weight corresponding to the architecture are obtained in the current training set. Compared with the prior art, the method has the advantage that the model with a better effect than manual design can be quickly found.

The invention discloses a time sequence physiological data classification method and device, a storage medium and a processor. The method comprises the following steps: extracting multisource sign data from a database, dividing the data into training data and test data, and preprocessing the training data and the test data; constructing a deep learning model DeepPhysioNet, wherein the model adoptsa coder-decoder neural network architecture, the coder is composed of a basic feature learning unit, a sequence residual error unit and a representation learning unit and can perform strong feature extraction, and the decoder calculates classification results for classification tasks of different targets by using extracted features; performing an offline training stage: inputting the training data into the model for preliminary training, testing the preliminarily trained model through the test data, and performing continuous repeating until a preset condition is met; performing the online inference stage: inputting to-be-detected data into the trained DeepPhysioNet model, and outputting a classification result. The method has the advantages that expert deviation is avoided, the method issuitable for multi-source time sequence physiological data, and an attention mechanism is introduced.